TechyCSRTechyCSR/OpenCluely/webapp/llms-full.txt

OpenCluely / webapp

llms.txt1.1k starsChanged 4 months agoApache-2.0View on GitHub

We split this llms-full.txt into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
No telemetry, no analytics, no auto-update beacon.
line 51
Distinctive:
Start your interview on any conferencing tool.
line 58
Distinctive:
Ask by voice (mic button or Alt+R) or by screenshot (Cmd/Ctrl+Shift+S).
line 59
Distinctive:
Do I need an API key? Yes, a free Google Gemini API key from Google AI Studio.
line 79

≈ “Common” is an estimate: at least 1 in 5 of the 3,225 llms.txt files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

llms-full.txt

# OpenCluely — extended AI guidance (llms-full.txt)
> A richer, structured description for AI crawlers and assistants.

## Project metadata

- Name: OpenCluely
- Alternate names: OpenCluely AI Interview Copilot, OpenCluely Stealth Overlay, OpenCluely Cluely Alternative
- Short description: Free, open-source AI interview copilot with a stealth overlay invisible to screen recording. MIT-licensed Cluely alternative.
- Long description: OpenCluely is a cross-platform Electron desktop application that overlays a small floating window on the user's screen during technical interviews. The overlay is excluded from screen capture by Zoom, Google Meet, Microsoft Teams, Discord, and OBS. The app accepts voice or screenshot input, sends it to Google Gemini for reasoning, and streams the answer back into the overlay word by word. It is a free MIT-licensed alternative to Cluely, Natively, Interview Coder, FinalRound AI, LockedIn AI, OffscreenAI, CodingVeil, Steal AI, and Ezzi. Architecture is model-agnostic by design; additional backends (OpenAI, Anthropic, local models) are planned.
- URL: https://opencluely.techycsr.dev/
- Repository: https://github.com/TechyCSR/OpenCluely
- Author: TechyCSR
- Author URL: https://techycsr.dev
- Maintainer: TechyCSR
- License: MIT (https://opensource.org/licenses/MIT)
- Created: 2025
- Current version: 1.0.0
- Last updated: 2026-07-29
- Pricing: Free (only user-paid Gemini API costs via BYOK)
- Privacy stance: No telemetry, no analytics, no data collection. All processing local except AI requests.
- Positioning: Free open-source alternative to paid AI interview copilots (Cluely, Natively, Interview Coder, FinalRound AI, LockedIn AI, OffscreenAI, CodingVeil, Steal AI, Ezzi).

## Platforms

- Windows 10/11 — NSIS .exe installer from GitHub Releases.
- Linux — .deb installer and AppImage from GitHub Releases. Tested on Ubuntu 20.04+, Debian 11+, Fedora 36+.
- macOS 12+ — Not pre-built (Gatekeeper blocks unsigned apps). Run from source with `./setup.sh`.

## Tech stack

- Runtime: Electron, Node.js.
- LLM backend: Google Gemini API (gemini-2.5-flash-lite, gemini-2.5-flash). BYOK model.
- Speech-to-text: OpenAI Whisper (local) with voice-activity detection, or Azure Speech Services.
- Vision: Gemini direct image input — no intermediate OCR step.
- Rendering: PrismJS for syntax highlighting, markdown for chat formatting.
- Window hiding: WDA_EXCLUDEFROMCAPTURE / SetWindowDisplayAffinity (Windows), NSWindowSharingNone (macOS).
- Build: GitHub Actions for Windows + Linux binaries.

## Features

- Stealth overlay invisible to screen recording (Zoom, Meet, Teams, Discord, OBS).
- Voice input with natural-pause voice activity detection (Whisper or Azure Speech).
- Screenshot input with direct Gemini vision analysis — works with HackerRank, LeetCode, CodeSignal, CoderPad, editors, slides, PDFs.
- Streamed answer rendering with markdown and code highlighting.
- Persistent session memory across the interview.
- Floating command bar and floating answer window, both click-through when not interacted with.
- System tray app for global shortcuts and quick toggles.
- Configurable keyboard shortcuts for every action.
- Settings UI for API keys, voice provider, model selection, and personalization.
- Multiple language support (Gemini-side).
- No telemetry, no analytics, no auto-update beacon.

## How to use

1. Install OpenCluely from the latest release on GitHub.
2. Launch the app. On first run the Settings window opens; paste a free Gemini API key from Google AI Studio.
3. Optionally configure Whisper (local, offline) or Azure Speech for voice input.
4. Start your interview on any conferencing tool. The overlay floats above your screen and stays hidden from screen recording.
5. Ask by voice (mic button or Alt+R) or by screenshot (Cmd/Ctrl+Shift+S). The answer streams in instantly.

## Keyboard shortcuts

| Shortcut | Action |
|---|---|
| Cmd/Ctrl+Shift+S | Screenshot capture |
| Alt+R | Toggle speech recognition |
| Cmd/Ctrl+Shift+V | Toggle window visibility |
| Cmd/Ctrl+Shift+I or Alt+A | Toggle click-through interaction |
| Cmd/Ctrl+Shift+C | Open chat |
| Cmd/Ctrl+, | Open settings |

## FAQ

- Is OpenCluely a free Cluely alternative? Yes — MIT licensed, no subscription, no account, no telemetry. BYOK keeps the cost near zero on Gemini's free tier.
- Is it really invisible during screen sharing? Yes on Windows and macOS, using OS-level window-exclusion flags (WDA_EXCLUDEFROMCAPTURE / SetWindowDisplayAffinity on Windows, NSWindowSharingNone on macOS). Linux lacks the API.
- How does it compare to Natively, Interview Coder, FinalRound AI, LockedIn AI? Same category. OpenCluely is the free MIT-licensed option with no subscription, no account, no telemetry, and cross-platform support (Windows, macOS, Linux).
- Does it support BYOK? Yes, BYOK is the only mode. Paste your own Gemini key.
- Does it work with HackerRank, LeetCode, CodeSignal? Yes — screenshot capture works in any application.
- Do I need an API key? Yes, a free Google Gemini API key from Google AI Studio.
- Is voice input required? No. Use screenshots and chat on their own.
- Is my data collected? No. Only the AI requests you send leave your device.
- Is it free? Yes, MIT licensed.
- Which platforms? Windows (installer), Linux (.deb, AppImage), macOS (source).
- Will it get me caught in an interview? Intended for learning, practice, and authorized use. Follow the rules of any interview you take.

## Related URLs

- Homepage: https://opencluely.techycsr.dev/
- GitHub: https://github.com/TechyCSR/OpenCluely
- Releases: https://github.com/TechyCSR/OpenCluely/releases/latest
- Releases feed (Atom): https://github.com/TechyCSR/OpenCluely/releases.atom
- Issue tracker: https://github.com/TechyCSR/OpenCluely/issues
- Developer: https://techycsr.dev
- License: https://opensource.org/licenses/MIT
- sitemap: https://opencluely.techycsr.dev/sitemap.xml
- robots: https://opencluely.techycsr.dev/robots.txt
- manifest: https://opencluely.techycsr.dev/manifest.webmanifest
- llms.txt: https://opencluely.techycsr.dev/llms.txt
- security.txt: https://opencluely.techycsr.dev/.well-known/security.txt
- humans.txt: https://opencluely.techycsr.dev/humans.txt
- og-image: https://opencluely.techycsr.dev/og-image.png

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free